Log in or Register for enhanced features | Forgotten Password?
White Papers | Suppliers | Events | Report Store | Companies | Dining Club | Videos
Design & Production
Design & Development
Return to: ABR Home | Design & Production | Design & Development

Continental uses AI to identify pedestrian intention and gestures

ABR Staff Writer Published 11 June 2018

German automotive supplier Continental is planning to use Artificial Intelligence (AI) and neural networks to help identify pedestrian intention and gesture classification for precise assessment of the entire traffic situation.

For an automated vehicle to assume control from drivers, the vehicle must develop an understanding of the imminent actions of all road users, so that they can make the right decision in different traffic situations.

Continental stated that this task can be performed by training algorithms leveraging deep machine learning methods. At the upcoming CES Asia, the company is all set to exhibit its computer vision platform that uses AI and neural networks alongside traditional computer vision processes.

The company said that its fifth generation of multi-function camera will enter into production in 2020 and will use neural networks alongside traditional computer vision processes.

This technology can be scaled and refined using intelligent algorithms they can improve the understanding of the intentions and gestures of pedestrians.

AI can open up possibilities for computer vision platform. AI can detect people and interpret their intentions and gestures. Continental claims that its AI can give vehicle systems human strengths.

A rule-based algorithm within an automated driving system can react when a pedestrian steps onto the road and detect the intention of the approaching pedestrian in advance. This is similar to the way in which experienced drivers react instinctively in such a situation and prepares to brake early.

These AI systems have to learn new skills, just like humans. To do this, the software will analyze huge amounts of data to derive successful and unsuccessful strategies for action and later apply their learned knowledge in the vehicle. This ability of algorithms to learn can be continually developed.

In advanced driver assistance systems, data for this kind of learning is available in the form of recorded radar and camera signals from real driving situations.

Continental advanced driver assistance system business unit head Karl Haupt said: “AI plays a large role in taking over human tasks. With AI software, the vehicle is able to interpret even complex and unforeseeable traffic situations – it’s no longer about what’s in front of me but about what could be in front of me. We see AI as a key technology for automated driving. AI is a part of an automotive future.”

Image: Continental’s AI at work in human gesture identification. Photo: Courtesy of Continental AG.